pytorch_vision_mobilenet_v2.md

layout	background-class	body-class	title	summary	category	image	author	tags	github-link	featured_image_1	featured_image_2
pytorch_hub_detail	pytorch-hub-background	pytorch-hub	MobileNet v2	The MobileNetV2 architecture is based on an inverted residual structure where the input and output of the residual block are thin bottleneck layers opposite to traditional residual models.	researchers	pytorch-logo.png	Pytorch Team	CV image classification	https://github.com/pytorch/vision.git	mobilenet_v2_1.png	mobilenet_v2_2.png

Model Description

The MobileNetV2 architecture is based on an inverted residual structure where the input and output of the residual block are thin bottleneck layers opposite to traditional residual models which use expanded representations in the input. MobileNetV2 uses lightweight depthwise convolutions to filter features in the intermediate expansion layer. Additionally, non-linearities in the narrow layers were removed in order to maintain representational power.

Model structure	Top-1 error	Top-5 error
mobilenet_v2	28.12	9.71

Notes on Inputs

All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape (3 x H x W), where H and W are expected to be at least 224. The images have to be loaded in to a range of [0, 1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225].

Example:

import torch
model = torch.hub.load('pytorch/vision', 'mobilenet_v2', pretrained=True)

Resources:

• MobileNetV2: Inverted Residuals and Linear Bottlenecks